Is this heat profile plausible? It's not at all what I expected. power temp 31 49 47 74 63 74 79 69 95 111 111 111 127 111 143 111 159 171 https://reprap.org/forum/read.php?14,5655,5655#msg-5655 Thu, 15 Oct 2026 10:34:18 -0400 Phorum 5.2.23 https://reprap.org/forum/read.php?14,5655,5684#msg-5684 https://reprap.org/forum/read.php?14,5655,5684#msg-5684 Karl Mechanics Mon, 15 Oct 2007 18:50:33 -0400 https://reprap.org/forum/read.php?14,5655,5662#msg-5662 https://reprap.org/forum/read.php?14,5655,5662#msg-5662 owad Mechanics Sun, 14 Oct 2007 18:30:48 -0400 https://reprap.org/forum/read.php?14,5655,5658#msg-5658 https://reprap.org/forum/read.php?14,5655,5658#msg-5658 Anonymous User Mechanics Sun, 14 Oct 2007 11:03:24 -0400 https://reprap.org/forum/read.php?14,5655,5656#msg-5656 https://reprap.org/forum/read.php?14,5655,5656#msg-5656
To measure temperature PIC measures the time it takes to charge a capacitor through the thermistor. When the temperature is higher, it takes less time to charge. As a result, the PIC can measure low temperatures with good precision, because the resistance is high, in the neighborhood of 10K. When you're over 100C, you're only seeing resistances of 1K - 400R or less, and the charge times get much faster, to the point where the smallest difference in charge time the PIC can measure corresponds to 20 or 30+ degrees.

The only way to get acceptable resolution in that range is to use a bigger capacitor. That can produce better readings at extruding temperatures, but it makes for much longer readings in the near-room-temperature range. Due to the way readings are done, longer readings fail more often, so you might never get valid readings at low temperatures.

As for solutions, I didn't find any way to get good results with the stock firmware and host software. Obviously people are using it and making stuff, I just haven't figured out how. The two solutions I've found are:

* use the stock firmware, put a fixed resistor in parallel with the thermistor to make its output more linear (suggested by nophead, documented on his blog), increase the size of the capacitor, and modify the host software to compensate for the parallel resistor. On paper, this sounds like it should work, but I got very poor results with it -- I kept getting invalid readings from the firmware. or,

* switch to the alternate firmware Andreas posted, and increase the size of the capacitor. This has been working very well for me -- I get accurate readings through the entire range of temperatures and no invalid readings.

IMHO, the second option is the way to go. Hopefully his firmware changes, or something similar, will make it into the standard firmware. Messing with the temperature readings has set me back a few weeks, and it looks like you're headed towards the same set of problems.]]> Anonymous User Mechanics Sat, 13 Oct 2007 23:31:40 -0400 https://reprap.org/forum/read.php?14,5655,5655#msg-5655 https://reprap.org/forum/read.php?14,5655,5655#msg-5655
power temp
31 49
47 74
63 74
79 69
95 111
111 111
127 111
143 111
159 171]]> owad Mechanics Sat, 13 Oct 2007 21:50:57 -0400